Autistic Traits and Sensitivity to Instruction

Background: Individuals with Autism Spectrum Disorder (ASD) and those high in autistic traits have been shown to be less influenced by contextual information accompanying the percept than controls (Jolliffe & Baron-Cohen, 1999; Noens & van Berckelaer-Onnes, 2008). Recently, Stewart & Ota (2008) investigated auditory speech perception using a Ganong paradigm; a segment identification test using two word-to-nonword Voice Onset Time (VOT) continua (kiss-giss and gift-kift). A significant negative correlation was found between autistic traits and the identification shift that occurred between the continua. As autistic traits increased participants were more likely to respond according to the actual acoustic difference, whereas those that had fewer autistic traits were more likely to respond in the direction of the real word. Pilot studies found variations in phoneme identification patterns as instructions differed.

Objectives: To investigate the extent, effect, and importance of emphasis in instructions when studying aspects of phonological processing and the relationship with autistic traits.

Methods: Twenty-four typically developing adults undertook a phoneme identification task designed to bias responses by providing contextual information immediately following the phoneme offset. Phonemes could be conceptualised as a seven point continuum from /gi/ to /ki/, with those items occurring near the midpoint being ambiguous. Instructions varied by emphasising either the lexical or auditory aspect of stimuli, such that participants’ attention would be drawn to those particular properties of the percept. Instructions with a lack of emphasis were also added as a control condition. All participants completed the Autism-Spectrum Quotient (AQ; Baron-Cohen, Wheelwright, Skinner, Martin & Clubley, 2001).

Results: When participants were asked to attend to the acoustic properties of the utterance, mean differences between continua were smaller, such that the lexical bias shown in the typical Ganong experiment was reduced. When asked to attend to the lexical properties, larger mean differences between continua suggested an amplified lexical effect as participants responded in favour of the word over the non-word. This instructional bias was correlated with scores on the AQ.

Conclusions: Marked differences in patterns of phoneme identification between conditions suggest that subtle differences in instruction can bias attentional focus when processing phonological stimuli. This appears to be enhanced in those who score more highly on the AQ, further suggesting that sensitivity to instruction may vary with autistic traits.